вход по аккаунту


Crystal Packing and Molecular Geometry.

код для вставкиСкачать
Crystal Packing and Molecular Geometry
J. Jens Wolff*
It is still not possible to deduce the packing of a molecular
crystal from the structure of its constituent molecules. Crystal
engineering, the often-stressed construction of crystal structures
at will, in general remains an unrealized vision. While great
strides have definitely been made since the pioneering work of
Kitaigorodsky, neither computational methods['] for the calculation of packing structures, nor experimental studies['] can
analyze this paradigm of molecular recognition in a completely
satisfactory manner. This comes as no surprise--crystal packing
is determined by a great number of factors, with the van der
Waals energy as the dominant term. Even quite smafl energy
differences may drastically alter a packing pattern. Parallels
may be drawn to stereoselective synthesis where small differences in energies, which are more than often the sum of different
effects, decide the success or failure of a particular reaction.
Recurring ordering motifs in the structures of molecular crystals
have at least been identified for crystals of aromatic hydrocarb o n ~ . In
~ ~this
" ~class of compounds, the possible intermolecular
interactions13b1are restricted to the van der Waals type, and the
molecular shape, since the occurrence of different conformers is
limited,f3c1restricts the number of possible packing patterns to
a few types (see Fig. 1). Even in this area, surprises are still
Fig. 1 Frequently occurring packing patterns for aromatic hydrocarbons.
hydrogen bonds as the organizing element.[51 The statistical
analysis of the contents of the Cambridge Crystallographic
Database@]has made it possible to identify a multitude of further intermolecular interactions in crystals like CH . . 10and
CI . . . CI contacts.[21However, it may be difficult to decide for a
given case to what extent these interactions really determine the
observed structure.
Special problems are presented by the phenomenon of polymorphi~rn,['~
which arises when one molecule crystallizes in two
or more geometrically different structures that are close in energy. Polymorphs may not even always be obtained reproducibly.[*] Preoccupation with the problem of polymorphism is
not necessarily esoteric.
For example, the acid
blocker ranitidine (2), the
M ~ , N & s - ~ ~ ~ ~ ~ ~
active constituent of the
best-selling drug worldwide in 1992, exists in two polymorphic forms of which only one
is bioaccessible. Several lawsuits against the manufacturers of
generic drugs, all won by the patent holder, centered around the
protectability of the active polymorph, which had been discovered seven years later than the inactive form."]
Clearly, the mode of molecular packing determines the practical value of a material not only in solid-state organic synthesesl'ol but also in all areas of applied research that depend on
the solid-state properties of molecular crystals. Many organic
pigments occur in different polymorphic forms that have different optical
(3), which for
in orange
to violet
possible as exemplified by the structural analysis of 1, which
unexpectedly crystallizes in a y-type rather than a sandwich
Structures with strongly directional intermolecular interactions like hydrogen bonds have also
been investigated f r e q ~ e n t l y . This
approach is akin to successful
supramolecular chemistry in solution which frequently draws upon
[*] Priv.-Doz. Dr. J. J. Wolff
Organisch-chemisches Institut der Universitdt
Im Neuenheimer Feld 270, D-69120 Heidelberg (Germany)
Fax. Int. code +(6221)544205
e-mail : wolff(ir
A n g w . Cliem. In!. Ed. Engl. 1996. 35, No. 19
modifications, and copper phthalo0
cyanine.'"] The use of crystalline
materials for second-order nonlinear optics['21 requires the
NLO-phores at least to crystallize in noncentrosymmetric space
a trivial problem to solve, because about threequarters of all nonchiral organic compounds adopt centric
space groups in the solid state." 31 Likewise, materials with special e I e c t r i ~ [ ' or
~ ] (ferro-)magnetic['51 properties require welldefined molecular arrangements.
Polymorphic forms often occur for molecules with high conformational mobility. Frequently in different polymorphs quite
different conformers are observed. At present, this interplay
between packing pattern on one hand and molecular conformation on the other is also little understood. Outside the realm of
crystallography, it is still common to consider one molecular
R? VCH Verlugsgesellschafr mbH. 0-69451 Weinheim. 1996
OS70-0833196/3S19-219S$ IS.OO+ .XI0
structure elucidated by diffraction methods as the structure of
the molecule; in other words, this structure is implicitly looked
upon as that of the nonperturbed, isolated molecule in the gas
phase. Molecular parameters that do not agree with this perception are thought to arise from “crystal packing forces”. Only in
rare cases may these forces be approximately quantified. Methods available for studying these crystal fields are: the comparison with data observed or computed for the gas-phase structure,
the comparison of crystallographically independent molecules
within the same crystal, and the comparison of molecular
parameters for different polymorphic crystal forms. Unfortunately, most molecules are either not volatile enough or too
complex in structure for sufficient information to be obtained
from spectroscopic methods in the gas phase. For many molecules only one crystal form with crystallographically identical
molecules is obtained (or studied). Four examples taken from
the recent literature (apart from the review articles cited) qualitatively illustrate the state of the art in this area, with its
successes as well as its problems: a systematic study of a class of
compounds and examples of structures distorted by ionic interactions, of conformational polymorphism, and of structuredependent photochromism.
For the study of the influence of crystal packing on molecular
parameters it is advantageous to use molecules with high polarizabilities and conformational mobilities. Recently, Glaser et al.
have tried to employ the azines
5 as probes for the quantification of crystal packing forces.[161
\ /
Schiff bases like 4 had already
proven valuable in the studies
of conformational polymorx--QfXN<N
phism by Bern~tein,~’~’
found crystal packing effects on
the order of 2 kJmol-‘ for the
rotamers that occur in the polymorphic forms of 4. Seventeen
azines of acetophenone (5)were investigated. The results show
that the gauche conformation around the N-N bond favored
for the gas-phase structure according to ab initio calculations
may be altered by packing effects. In some crystals the frans
arrangement is found. However, the expectation of push -pull
effects on the molecular conformation (e.g. for X = M e 0 and
Y = CN) could not be substantiated, since the molecular
parameters of symmetrically substituted compounds do not differ appreciably from those with push -pull sub-stitution. Likewise, the crystal packing effects observed could be explained solely on the basis of van der Waals contributions to the packing
energy without invoking any polar contributions.
Dahne and Reek["] succeeded in distorting the equalized
C-C bond lengths along the chain of the cyanine 6 in the solid
state to a maximum alternation
Me2N-. -.
._I ._*- -..-.
of 0.02 A, an effect that had not
been encountered before with
small counterions like tetrafluoroborate. The sterically demanding tetraphenylborate ion
requires a good van der Waals contact for optimum packing. As
a consequence, the distribution of negative charge in the vicinity
ofthe cyanine cation becomes quite unsymmetrical, which leads
to a much greater distortion as would be expected for a corresponding point charge.
Conformational polymorphism was
observed with 7.[18a1
Two different ro7
tamers with respect to the C-N bond
are found in two polymorphs which,
due to the different ability for conjugation with the a-system, also differ in the pyramidality at the nitrogen atom and
in the @so angles of the benzene rings. PM3 calculations indicated that the conformers are very close in
energy; the differences in the packing energies
between the polymorphs fall into the range
of about 4 kJmol-’. Similar effects on the
molecular structure had been found before for
8 for which two C-N rotamers coexist in
the same crystal.[18b1
Crystals of 9, a member of a class of compounds known to be
photochromic, exist in two polymorphic forms of which only
one shows photochromic behavior.[”] The arrangement of 9 in
the polymorphs shows distinct differences. The inactivity of the
first polymorph was ascribed to the short intermolecular contacts between the molecules, which could lead to the deactivation of the excited state and thus impede the proton transfer.
This study also highlights the far-reaching consequences that
even extremely subtle changes in molecular shape may have on
crystallization: even deuteration of the methylene group of 9
markedly changes the ratio in which the two polymorphs occur.
Especially this last observation could leave doubts whether an
ab initio description of the packing of organic molecular crystals
will ever be possible. Certainly, a precise quantificiation of the
on the molecular conformation will
influence of crystal
be very useful. On the other hand, the nucleation and the kinetics of growth of molecular crystals[”] will have to be studied
intensely, for example by observation of the integration of contaminants into the crystal lattice; optically anomalous crystalsrzb]are suitable candidates for studies of this kind. Reducing
the problem of crystallization from three to two dimensions,
as offered by the investigation of the self-organization of
monomolecular layers on surfaces, also promises decisive progress.
German version: Angew. Chem. 1996, 108, 2339-2341
Keywords: crystal engineering packing forces - polymorphism
* structure predictions
Verlagsgesellschaft mbH. 0-69451 Weinheim, 1996
[I] a) A. Kitaigorodsky, Molecular Crysfalsand Molecules, Academic Press, New
York, 1973; b) A. Gavezzotti, J. Am. Chem. Soc. 1991,113,4622; Acc. Chem.
Comput. Chem. 1992,
Res. 1994, 27, 309; c) H. R. Karfunkel, R. J. Gdanitz, .
13, 1171; d) A. M. Chaka, R. Zaniewski, W. Youngs, C. Tessier, G. Klopman,
Acta Crysfallogr.Serf. B 1996.52, 165. e) For prediction of polymorphs: D. E.
Williams, Acta Cr.ysfallogr.Secf. A 1996, S2. 326. f) More recent search algorithms improve the efficiency of the procedures substantially: P. Erk, Ludwigshafen, private communication. Computational methods are of great importance for structure determinations from powder diagrams.
[2] a) G. R. Desiraju, Angew. Chem. 1995,107.2541;Angew. Chem. I n f . Ed. Engl.
1995, 34, 2328. The parallels drawn there between organic synthesis and the
synthesis of crystal structures may also be found in: b) B. Kahr, J. M. McBride,
0570-0833196j3519-2196 8 1S.OOf .25/0
Angew. Chem. Int. Ed. En@. 1996. 35. No. 19
Angels. Chem. 1992, 104. 1 ; Angew. Chem. Int. Ed. Engl. 1992.31, 1 . c) For the
influence of weak hydrogen bonds on crystal structures, see also: T. Steiner, W.
Saenger, J Chem. Sac. Chem. Commun. 1995,2087.
131 a) G. R. Desiraju. A. Gavezzotti. Acta Crystallogr. Sect. B 1989. 45,473. The
similarities in the packing of aromatic molecules are also demonstrated by the
success of increment systems to compute thermodynamic data: K. Nass, D.
Lenoir. A. Kettrup, Angew Chem. 1995. 107, 1865; Angew. Chem. Int. Ed.
Engl. 1995,34. 1735. b) For the well-understood interaction (molecular recognition) between aromatic rr-systems, see: C. A. Hunter, Chem. Soc. Rev. 1994,
23.101 ;A. W. Schwabacher, 2. Shuhog. W. Davy, J Am. Chem. Sac. 1993.1 I S ,
6995; F. Cozzi, F. Ponzini, R. Annunziata. M. Cinquini, J. S . Siege], Angew.
Chem 1995. 107, 1092; Angew. Chem. I n [ . Ed. Engl. 1995. 34, 1019; R.
Laatikainen. J. Ratilainen, R. Sebastran, H. Santa, J. Am. Chem. Sac. 1995,
117. 11006. c) Even {he introduction of one degree of rotational freedom, as in
the polyphenyls. leads to crystallographically unpleasant problems like incommensurate phases: J. L. Baudour, Actu Crystullogr. Secr B 1991.47.935, and
references therein; d) R. Goddard, M. W. Haenel, W. C. Herndon, C. Kriiger,
M . Zander, J. Am. Chem. SOC.1995, 117, 30, and references therein.
[4] a ) J. Bernstein. R. E. Davis. L. Shimoni, N.-L. Chang. Angew. Chem. 1995,107,
1689: Angel“. Chem. I n f . Ed. Engi. 1995, 34, 1545; b) A. D. Burrows, C.-W.
Chan. M. M. Chowdhry, J. E. McGrady, D. M. P. Mingos, Chem. Sac. Rev.
1995,24.329:C. B. Aakeroy, K. R. Seddon, ihid 1993,22,397;c) M. C. Etter.
S. M. Reutzel. J. Am. Chem. Sac. 1991,113,2586;M. C. Etter, Acr. Chem. Res.
1990. 23. 120.
IS] H. Schneider. Angew Chem. 1991. 103, 1419; Angen. Chem. In[. Ed. Engl.
1991. 30. 1417: G. M. Whitesides. E. E. Simanek, J. P. Mathias, C. T. Seto,
D. N. Chin. M. Mammen, D. M. Gordon, Acc. Chem. Res. 1995, 28, 37; E.
Fan. J Yang. S. J. Geib. T. C. Stoner, M. D. Hopkins. A. D. Hamilton, J
Chem. So<. Cheni. Commun. 1995, 251.
[6] a ) Stru<turc. Correlation (Eds.: H:B. Biirgi. J. D. Dunitz), VCH. Weinheim,
1994; b) F. H. Allen, 0. Kennard, D. G. Watson in [6a], pp. 71-110; c) R.
Taylor. F. H. Allen. in [6a]. pp. 111-161.
[7] a) One of the pioneers in this area was Hantzsch, whose observations on
“chromoisomerism”. that is, polymorphs with different spectral properties, are
still frequently perused (though rarely cited): A. Hantzsch, Justus Liebigs Ann.
Chem. 1931. 492. 65; b) A. Gavezzotti. G. Filippini, J Am. Chem. Sac. 1995,
117. 12299: c ) J. D. Dunitz. Pure Appl. Chem. 1991,63, 177; Acta Crystallogr.
Sucr. B. 1995. 51, 619; d) I . Bar, J. Bernstein, Tetrahedron 1987, 43, 1299; J.
Bernstein, in Accurate Moleculur Structures (Eds.: A. Domenicano, I. Hargittai). Oxford University Press, New York, 1992, pp. 469-497; J. Bernstein, in
Orgunrc Solrd State Chemistry (Ed.: G. R. Desiraju) Elsevier, Amsterdam,
1987. pp- 471 -518; e) J. H. Matthews. I. C. Paul, D. Y Curtin, J. Chem. Sac.
Perkin Trans 2 1991. 113; D. Y. Curtin, I. C. Paul. Chem. Rev. 1981, 81, 525:
f ) Cosolvate- E. Weber, S . Finge, I. Csoregh, 1.Org. Chem. 1991,56,7281;Top.
Curr. Chum. 1987, 140 and 1988, 149. h) For crystallization in general, see: J.
Hulliger. Angov. Chem. 1994, 106, 151 ;Angen. Chem. I n t . Ed. Engl. 1994, 33,
[8] Several CdSeS of (seemingly) disappearing polymorphs are discussed by: J. D.
Dunitz. J. Bernsrein. A x Chem. Res. 1995, 28. 193.
Aneew. Chern. In/. Ed. Engl.
1996, 35; No. 19
[9] C. Leadbeater, Financial Times, April 9, 1991.
[lo] Cf. references in: J. J. Wolff, Angew. Chem. 1995.107,2407. Angew. Chem. Inr.
Ed. Engl. 1995.34.222s. For some recent cases, see also: F. Toda, Acc. Chem.
Res. 1995.28. 480; M. SmrEina, VyskoEil, V. HanuS, M. PolaSek, V. Langer,
B. G. M. Chew. D. B. Zax. H. Verrier, K. Harper, T. A. Claxton, P. KoEovsky,
J. Am. Chem. Sac. 1996, 118,487.
[ l l ] W. Herbst, K. Hunger, Industrrelle organische Pigmente, 2nd vol., VCH. Weinheim, 1995, p. 468; H. Zollinger, Color Chembfry. 2nd ed., VCH, Weinheim,
1991, p. 240ff. A recent example taken from the area of organic dyes is given
by: S. J. Bonafede, M. D. Ward. J. Am. Chem. SOC.1995. / 1 7 , 7853.
[12] Special issue devoted to NL0:Chem. Rev. 1994. 94, 1-278. Nonlinear Opticul
Properties of Organic Molecules and Cr.vstals (Eds.: D. S. Chemla. J. Zyss,
Academic Press, Orlando. FL (USA), 1987.
[13] For the problem of centrosymmetric space groups, see: A. J. C. Wilson, Acta
Cr.vsta1logr. Sect. A 1990, 46. 742; G. R. Desiraju, Cr-vstal Engineering-The
Design of Organic Soldy, Materials Science Monographs 54. Elsevier, Amsterdam, 1989, Chapter 8; C. P. Brock, J. D. Dunitz, Chem. Muter. 1994, 6. 1118.
For the correlation between molecular dipole moment and the formation of
centric space groups, see: J. K. Whitesell, R. E. Davis. L. L. Saunders. R. J.
Wilson, J. P. Fedgins, 1. Am. Chem. Sac. 1991, 113, 3267.
[14] M. R. Bryce, Chem. So(. Rev. 1991,20,355; F. J. Adrian. D. 0 .Cowan, Chem.
Eng. News 1992. 70(51), 24.
[IS] J. S. Miller, A. J. Epstein, Angew. Chem. 1994, 106.399;Angew. Chem. Int. Ed.
Engl. 1994, 33, 385.
1161 G. Shiahuy Chen, J. K. Wilbur, C . L. Barnes, R. Glaser, J. Chem. Sac. Perkin
Trans 2 1995, 2311, and references therein. The criteria of high polarizability
and conformational lability are also fulfilled by benzene derivatives that are
threefold substituted with strong donors and acceptors which tend to form
cosolvates (pseudopolymorphs): J. J. Wolff, H. Irngartinger, F. Gredel, I. Bolocan, Chem. Ber. 1993. 126. 2127.
1171 L. Dahne, G. Reck, Angew. Chem. 1995,107,735;Angew. Chem. Int. Ed. Engl.
1995. 34. 690. For differences in N = N bond lengths that arise through the
coordination of boron-containing anions, see the analysis of a diazenium
salt with eight crystallographically independent molecules in the unit cell:
S. F. Nelsen, H. Chang, J. J. Wolff, D. R. Powell, J. Org. Chem. 1994, 59,
[IS] a) H. Bock, I. Gobel. C. Nather, Z. Havlas, A. Gavezzotti, G. Filippini, Angew.
Chem. 1993. 105. 1823; Angew. Chem. I n [ . Ed. Engl. 1993. 32, 1755; b) M
Kaftory, in Acyclic Organonitragen Stereodynamics (Eds.: J. B. Lambert. Y.
Takeuchi). VCH, New York, 1992, p. 263; A. Domenicano, A. Vaciago, Acta
Crystallogr. Sect. B 1919. 35, 1382.
1191 Y. Eichen. J.-M. Lehn, M. Schwerl, D. Haarer, J. Fischer. A. DeCian, A.
Corval, H. P. Trommsdorff, Angew. Chem. 1995,107,2753:Angew. Chem. Int.
Ed. Engl. 1995, 34, 2530.
[201 For theoretical strategies to model the electrostatic influence on the molecular
conformations in crystals, see, for example: P. Popelier. A. T. H. Leustra. C.
van Alsenoy, H. J. Geise, J. Am. Chem. Sac. 1989, I l l , 5658.
[21] I. Weissbuch, L. Addadi, M. Lahav, L. Leiserowitz. Science 1991, 253, 637.
VCH Verlagsgesellschaft mbH, D-69451 Weinherm, 1996
OS70-0833/96/3519-21973 I5.00+.2S/O
Без категории
Размер файла
352 Кб
geometry, crystals, packing, molecular
Пожаловаться на содержимое документа